As a first step in this analysis, a strategy to make 
point mutations in every repeat of the CTD has been 
developed. This was achieved by reconstructing the 
CTD coding sequence from synthetic oligonucleo- 
tides coding for either wild-type or mutant repeats. 
These synthetic CTDs are then cloned into a spe- 
cially constructed polymerase expression vector. 
Several types of CTDs have been constructed and 
analyzed. First, deletion studies confirmed the re- 
sults of previous experiments showing that CTDs 
with more than 1 1 of the normal 26 repeats behave 
identically to wild type. The minimum viable CTD 
is eight repeats long, and this strain exhibits a severe 
slow-growth phenotype. 
The expression of some inducible genes has been 
shown to be reduced in short CTD strains. This can- 
not, however, explain the slow growth seen in rich 
media. In Dr. Corden's laboratory, clones of several 
genes underexpressed in the short CTD strain were 
identified, using differential screening. All five dif- 
ferentially reduced genes identified in the screen 
are glycolytic enzymes. 
Based on this initial observation, the expression of 
other glycolytic enzymes was tested and shown to be 
reduced in a short CTD strain. The enzymes of the 
glycolytic pathway make up about 60% of the cyto- 
solic protein in yeast and therefore represent a large 
percentage of transcription initiation events in the 
cell. However, the three- to fivefold reduction seen 
in the short CTD strain is not simply a reduction in 
the most highly transcribed genes, as several highly 
expressed genes that do not encode glycolytic en- 
zymes are not reduced. Rather, the CTD is probably 
involved somehow in the specific expression of the 
genes encoding glycolytic enzymes. These genes 
have several transcription factors in common, and 
current studies are aimed at determining the role of 
the CTD in their action. 
Another important finding is that RNA polymerase 
II that lacks the CTD has a dominant negative pheno- 
type. Photoafl&nity labeling experiments indicate 
that this enzyme is incapable of transcribing in vivo. 
Its ability to interfere with transcription by the wild- 
type enzyme might, therefore, be due to competi- 
tion at the level of formation of the initiation com- 
plex. If the CTD-less enzyme can bind the promoter 
but not initiate transcription, then it may interfere 
with binding and initiation of the wild-type enzyme. 
Current in vitro studies are designed to test this in- 
terference hypothesis. 
Mutations that lengthen the CTD have just as seri- 
ous a consequence as the short CTD mutants. In 
addition to growing slower than their wild-type 
counterparts, the long-tail subunit is aberrantly 
phosphorylated in yeast. One suppressor of the 
long-tail mutant maps to the GRRl gene, which 
was previously shown to be involved in protein de- 
phosphorylation. Whether this protein plays a 
role in transcriptional regulation is now under 
investigation. 
Actively-transcribing RNA polymerase II is known 
to be highly phosphorylated on the CTD, and this 
modification is thought to take place during the ini- 
tiation reaction. Dr. Corden's laboratory previously 
identified several protein kinases that are able to 
phosphorylate the CTD at serines in positions two 
and five. These residues have been mutated to ala- 
nine or glutamate, and the eff'ect of these changes 
has been shown to be lethal. This observation is con- 
sistent with a requirement for phosphorylation of 
these residues in vivo. 
Genetic suppression of lethal CTD mutations is 
being developed as a method for identifying pro- 
teins that interact with the CTD. Suppressors have 
been isolated that allow the growth of mutations in 
the CTD that would otherwise be lethal. A collec- 
tion of 60 different strains containing suppressors of 
four different phosphorylation-site mutations are 
currently being characterized. At least one of these 
suppressors seems not to support growth in the pres- 
ence of wild-type CTD, indicating a potential 
change in specificity of an enzyme or other protein 
that interacts with the CTD. 
Previous studies in Dr. Corden's laboratory indi- 
cated that the CTD of mammalian species is highly 
conserved through evolution. This observation was 
unexpected in light of genetic studies showing that 
nearly half of the mouse CTD can be deleted with- 
out impairing the growth of cells in culture. Dr. 
Corden's laboratory is currently testing the hypoth- 
esis that the CTD plays a role in transcription during 
early development. 
The laboratory has also continued a collaboration 
with Drs. Terry Brown and Claude Migeon of the 
Johns Hopkins University School of Medicine on the 
function of the human androgen receptor (AR) . Dr. 
Corden's laboratory is currently investigating the 
mechanism by which the AR interacts with the basal 
transcription machinery to activate transcription of 
androgen-responsive genes. 
Dr. Corden is also Associate Professor of Molec- 
ular Biology and Genetics at the fohns Hopkins 
University School of Medicine. 
Books and Chapters of Books 
Corden, J.L., and Ingles, C.J. 1992. Carboxy- 
terminal domain of the largest subunit of eukary- 
otic RNA polymerase II. In Transcriptional Regu- 
lation (McKnight, S.L., and Yamamoto, K., 
176 
